Qwen 3 235B-A22B

Name: Qwen 3 235B-A22B
Rating: 4.5 (1 reviews)
Author: Alibaba

Open Source

By AlibabaReleased 2025-04-29128K context235B (22B active) parameters

Open SourceAPI Available

Benchmark Scores

89.4/100

MMLU89.5

HumanEval91.5

Math85

Reasoning90.2

Coding91

About Qwen 3 235B-A22B

Qwen 3 235B-A22B is Alibaba's flagship MoE model with thinking capabilities that compete with DeepSeek R1 and other frontier reasoning models. It uses 22B active parameters from a 235B total pool.

Strengths

+Competitive with DeepSeek R1
+MoE efficiency
+Strong thinking mode
+Open source

Weaknesses

-Newer model with limited ecosystem
-Complex deployment
-Variable quality across tasks

Pricing

Per 1M tokens

Input$0.40

Output$1.60

Quick Facts

Context Window: 128K
Parameters: 235B (22B active)
Release Date: 2025-04-29
Category: Open Source

Compare Models

See how Qwen 3 235B-A22B stacks up against other models